Atrial fibrillation (AF) is a cardiac rhythm abnormality that currently affects over 6 million Americans. This statistic is expected to double over the next decade given the increasing prevalence of AF risk factors such as advanced age and obesity. Atrial fibrillation confers a 5-fold risk of ischemic stroke, but can be treated effectively with anticoagulation therapy. Despite the efficacy of available treatment options, 1 in 5 patients with AF present with stroke as their initial manifestation of the arrhythmia. This is attributable to the significant challenge in diagnosing AF due to its episodic and sometimes asymptomatic nature. Existing AF monitoring strategies are burdensome or costly and invasive, and thus have low patient adherence and satisfaction. Recently, commercially available wrist-based wearable devices, or smartwatches, have shown to be accurate for AF detection, and may represent a promising tool for identifying AF. However, commercial devices are not primarily designed for use by older adults for arrhythmia detection, and there is a significant research gap in the feasibility of using smartwatches for arrhythmia detection in this population. Furthermore, no previous research has investigated the potential for implementation and integration of smartwatches into the healthcare system and infrastructure. Using data collected from the in-house randomized control trial Pulsewatch, and by conducting qualitative assessments in usability and implementation, this proposal addresses the evidence gap in the feasibility of smartwatches for AF detection with three specific aims: 1) to evaluate individual-level factors associated with adherence of using a smartwatch for AF detection, 2) to explore patient characteristics associated with acceptability of smartwatches and identify specific usability challenges and nuances for older adults, and 3) to identify barriers and facilitators of implementing smartwatches for use in a clinical setting. We approach the problem with a user-centered focus and apply rigorous and systematic scientific methods in completing these aims. Knowledge generated from this proposal will provide future researchers and stakeholders with practical evidence in the potential use of smartwatches for detection of AF.